Agent skill

Mg CLI

by modelguide in modelguide/modelguide

Generate YAML configuration files and run CLI commands to onboard organizations into ModelGuide.

MITAuto-check passedAI & LLM Engineering

Install Mg CLI

skills CLI
$ npx skills add modelguide/modelguide --skill mg-cli -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install modelguide/modelguide mg-cli --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/modelguide/modelguide.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/mg-cli .claude/skills/mg-cli && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
mg-cli
GitHub stars
108
Token cost
~3.5k tokens
SKILL.md length
663 words
Files
4 (incl. references)
Skills in repo
7
Repo updated
First seen
Licence
MIT

At a glance

Generate YAML configuration files and run CLI commands to onboard organizations into ModelGuide.

  • Works in 5 steps: Gather requirements: What connectors do… → Create the directory: anywhere the user… → Write files in dependency order: org →… → …
  • The user asks to set up an org
  • SKILL.md covers When to Use This Skill, Workflow Overview, Pipeline Dependency Order and Idempotency, plus 4 more sections
  • Calls bun and railway; reaches store.acme.com; needs APP_DB_PASSWORD

What it does

Mg CLI is an agent skill from modelguide/modelguide. Generate YAML configuration files and run CLI commands to onboard organizations into ModelGuide. Use this skill when the user asks to set up an org, create agents, import SOPs, add connectors, prepare onboarding YAML, provision a customer, seed demo data, or anything related to the mg CLI tool. Also trigger when the user mentions "mg setup", "mg import", "mg add", "onboard", "provision org", "create YAML for CLI", "prepare config files", or asks how to get a new organization running in ModelGuide.

Its SKILL.md is about 3.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `references/catalog.md`, `references/examples.md` and `references/schemas.md`).

It sits in AI & LLM Engineering, covering Operations and SOPs and Building AI agents. The repository describes itself as: Open-source voice agent orchestration framework - build production voice AI pipelines without vendor lock-in. The licence is MIT.

When your agent uses it

  • The user asks to set up an org
  • Prepare onboarding YAML
  • Provision a customer
  • Anything related to the mg CLI tool

Example prompts

  • “mg setup”
  • “mg import”
  • “mg add”
  • “/mg-cli”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Gather requirements: What connectors do they need? What agents? What workflows (SOPs)?
  2. Create the directory: anywhere the user wants (e.g., ~/onboarding/my-customer/)
  3. Write files in dependency order: org → users → secrets → connectors → agents → sops → guardrails → evals → sessions
  4. Validate with dry-run: cd modelguide-api && bun run src/cli/mg.ts setup /path/to/dir --dry-run
  5. Run the import: add --skip-secrets for testing, or run without flags for production

What it can do on your machine

Read from SKILL.md and the folder at commit 554caa0. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • bun
    • railway

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • store.acme.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • APP_DB_PASSWORD

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Mg CLI loads about 3.5k tokens when it runs, and up to ~13k if it reads all its reference files. Until then it costs about 128 tokens; SKILL.md has 663 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~128
When it runs · the whole SKILL.md, loaded when a task matches
~3.5k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~13k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from modelguide/modelguide at commit 554caa0, republished under its MIT licence (© modelguide). 663 words, ~3,471 tokens.

Download SKILL.mdSave it as .claude/skills/mg-cli/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
mg-cli
description
Generate YAML configuration files and run CLI commands to onboard organizations into ModelGuide. Use this skill when the user asks to set up an org, create agents, import SOPs, add connectors, prepare onboarding YAML, provision a customer, seed demo data, or anything related to the `mg` CLI tool. Also trigger when the user mentions "mg setup", "mg import", "mg add", "onboard", "provision org", "create YAML for CLI", "prepare config files", or asks how to get a new organization running in ModelGuide.

ModelGuide CLI Onboarding Tool

The mg CLI is a thin orchestration layer over ModelGuide's service layer. It reads YAML files, validates them with Zod, and calls existing @features/* services in dependency order. All business logic lives in the services — the CLI handles parsing, validation, orchestration, and output.

When to Use This Skill

  • User wants to onboard a new organization (customer, demo, test)
  • User wants to prepare YAML config files for the CLI
  • User wants to run individual CLI commands or the full mg setup pipeline
  • User wants to import SOPs, guardrails, evals, sessions, or other entities
  • User asks about available connectors, SOP templates, or schema fields

Workflow Overview

There are two ways to use the CLI:

Create a directory with YAML files anywhere on disk, then run one command:

bash
cd modelguide-api
bun run src/cli/mg.ts setup /path/to/my-org/           # provision everything
bun run src/cli/mg.ts setup /path/to/my-org/ --dry-run  # validate and preview without changes

The directory can live anywhere — it does not need to be inside the modelguide repo. Pass an absolute or relative path.

Required file: org.yaml Optional files: users.yaml, secrets.yaml, connectors.yaml, agents.yaml, sops.yaml, guardrails.yaml, evals.yaml (or evals-*.yaml for multi-agent orgs), sessions.yaml

Flags:

  • --dry-run — validate all YAML files against schemas and print the plan without touching the database. Use this to verify files are correct before running for real.
  • --skip-secrets — use placeholder values (useful for CI/testing)
  • --skip-compile — skip agent compilation step
  • --skip-evals — skip eval import
  • --skip-sessions — skip demo session import
Option B: Individual Commands

Run each step separately (useful for adding to an existing org):

bash
cd modelguide-api
bun run src/cli/mg.ts create-org --from /path/to/org.yaml
bun run src/cli/mg.ts add-users --org acme --from /path/to/users.yaml
bun run src/cli/mg.ts add-secrets --org acme --from /path/to/secrets.yaml
bun run src/cli/mg.ts add-connectors --org acme --from /path/to/connectors.yaml
bun run src/cli/mg.ts add-agents --org acme --from /path/to/agents.yaml
bun run src/cli/mg.ts import-sops --org acme /path/to/sops.yaml
bun run src/cli/mg.ts import-guardrails --org acme /path/to/guardrails.yaml
bun run src/cli/mg.ts import-evals --org acme /path/to/evals.yaml
bun run src/cli/mg.ts compile-agents --org acme
bun run src/cli/mg.ts import-sessions --org acme /path/to/sessions.yaml

Pipeline Dependency Order

The order matters because later steps reference entities created earlier:

1. org.yaml        — organization (everything scoped to this)
2. users.yaml      — users (agents need a createdBy user)
3. secrets.yaml    — standalone secrets (connectors may reference these)
4. connectors.yaml — connectors + connector-scoped secrets
5. agents.yaml     — agents + tool assignments (references connectors)
6. sops.yaml       — SOPs (references agents + connector tools)
7. guardrails.yaml — guardrails (references agents)
8. evals.yaml     — eval suites, evaluators, test cases (references agents + SOPs)
9. compile-agents  — compiles each agent against its active SOPs (skipped with --skip-compile)
10. sessions.yaml  — demo sessions (references agents)

The mg setup command handles this order automatically and threads an IdRegistry (slug-to-UUID map) across all steps so cross-references resolve without extra DB queries.

Idempotency

Re-running is safe:

  • Orgs: upsert on slug (updates settings if exists, warns)
  • Users/Agents/Connectors/SOPs/Guardrails: duplicate errors are caught and counted as "existing"
  • Evals: suites deduped by (agent, SOP) pair; test cases by externalId in JSONB; eval configs by name
  • Sessions: deduped by externalId (explicit or derived from payload hash)
  • Secrets: append-only (no stable dedup key — use --skip-secrets on re-runs)

Quick Schema Reference

Each YAML file has a specific structure. For the complete field-by-field reference with types, defaults, constraints, and edge cases, read references/schemas.md.

org.yaml
yaml
name: "Acme Corp"
slug: "acme"                    # lowercase + hyphens only
timezone: "America/Chicago"     # optional
features: [voice-agents]        # optional
demoEnabled: false              # optional, default false
users.yaml
yaml
users:
  - email: admin@acme.example.com
    name: "Alice Admin"
    role: admin                 # admin | support
secrets.yaml
yaml
secrets:
  - name: OpenAI API Key
    type: platform_api_key      # api_key | oauth_token | credentials | platform_api_key | webhook_secret
    scope: agent                # connector | agent (optional)
    # value: omitted = prompted interactively (or placeholder with --skip-secrets)
connectors.yaml
yaml
# Real connector — references a registered TypeScript manifest
connectors:
  - name: "Acme Store"
    slug: "acme_store"          # lowercase + underscores
    catalogSlug: "medusa"       # must match a catalog entry — see references/catalog.md
    config:
      baseUrl: "https://api.acme.example.com"
    secrets:                    # connector-scoped secrets created automatically
      - field: "secretApiKey"    # field name in connector config
        name: "Acme Store API Key"
        type: api_key

  # Mocked connector — DB-driven fixtures, no TypeScript handler (ADR-013)
  - name: "Bank Nowa Banking (Mock)"
    slug: "banknowa_banking"
    isMocked: true              # switches schema branch
    iconUrl: "/logos/bank-nowa.svg"   # optional
    tools:                       # inline tool defs — each returns `mock_response` verbatim
      - name: "Verify Customer"
        description: "Verify identity."
        input_schema:
          type: object
          properties: { name: {type: string} }
          required: [name]
        mock_response:
          success: true
          customer_id: "CUST-001"

Edit mock_response in YAML and re-run mg add-connectors — existing tool rows are reconciled (no delete-then-reimport needed). See references/schemas.md for full field tables.

agents.yaml
yaml
agents:
  - name: "Acme Voice Agent"
    slug: "acme-voice-agent"
    description: "Handles phone orders"
    modality: voice             # voice | text (default: voice)
    platform: custom            # custom | elevenlabs | livekit (default: custom)
    tools:
      - connectorSlug: "acme_store"           # all tools from this connector
      - connectorSlug: "acme_support"
        toolSlugs: [create_ticket, get_ticket] # specific tools only

For platform: livekit (voice-test + outbound dispatch require this):

yaml
agents:
  - name: "Acme Voice Agent"
    slug: "acme-voice-agent"
    modality: voice
    platform: livekit
    config:
      # Only url + agentName are valid for livekit. llmModel is rejected
      # (baked into the worker image).
      url: "wss://your-project.livekit.cloud"
      agentName: "acme_voice_agent"   # must match the profile key in the worker's config/agents.yaml
    tools:
      - connectorSlug: "acme_store"
    secrets:
      # No `value:` → `mg setup` prompts once per field. These exact field
      # names are read by agents.service.ts:getAgentSecretByType when
      # dispatching the worker.
      - field: livekit_api_key
        name: "LiveKit API Key"
        type: api_key
      - field: livekit_api_secret
        name: "LiveKit API Secret"
        type: api_key
Show full SKILL.md (269 more words)Show less
sops.yaml — two modes

Inline SOP (define steps directly):

yaml
sops:
  - name: "Order Lookup"
    slug: "order-lookup"
    status: active              # draft | active | archived (default: draft)
    agents: ["acme-voice-agent"]
    trigger:
      type: intent_detected          # see references/schemas.md for all trigger types
      config:
        patterns: ["where is my order", "track my order", "order status"]
    steps:
      - id: greet
        instruction: "Greet and ask for order number"
        required: true
      - id: lookup
        instruction: "Look up the order"
        required: true
        tool:
          connectorSlug: "acme_store"
          toolSlug: "get_order"

Template fork (fork from a global SOP template):

yaml
sops:
  - name: "Order Lookup"
    templateSlug: "order-lookup"   # must match a template — see references/catalog.md
    status: active
    agents: ["acme-voice-agent"]
    connectorMapping:
      medusa: "acme_store"         # maps template's catalog refs to org's connector slugs

Cannot specify both templateSlug and steps.

guardrails.yaml
yaml
guardrails:
  - name: "No Medical Claims"
    slug: "no-medical-claims"
    content: |
      Never claim any product treats, cures, or prevents a medical condition.
    description: "FDA compliance"
    config: { priority: critical, category: compliance }
    agents: ["acme-voice-agent", "acme-chat-assistant"]
evals.yaml

One file per agent. For multi-agent orgs, use multiple files: evals-insurance.yaml, evals-booking.yaml, etc. The mg setup pipeline globs for evals*.yaml.

yaml
agentSlug: acme-voice-agent

evaluators:
  - name: confirms-order-id
    criterion: Agent confirms the order ID back to the customer
    tags: [accuracy]                    # optional
  - name: does-not-fabricate
    criterion: Agent does NOT make up order details or tracking information
    tags: [compliance, accuracy]        # optional

test_cases:
  - id: order-lookup-happy-path-01
    sop_slug: order-lookup
    scenario_key: order_status          # optional
    tags: [order-lookup, happy-path]    # optional
    evaluators:                         # references by name
      - confirms-order-id
      - does-not-fabricate
    input:
      customer_message: Hi, I placed an order last week, number ACM-12345.
      conversation_history:
        - role: assistant
          content: Thanks for calling Acme Corp. How can I help you today?

Also supports standalone import via JSON (eval-scenarios.json) with --agent flag:

bash
bun run src/cli/mg.ts import-evals --org acme --agent acme-voice-agent /path/to/eval-scenarios.json
sessions.yaml
yaml
sessions:
  - agentSlug: "acme-voice-agent"
    channel: voice              # voice | web | api | slack | widget | sms | whatsapp | email
    status: completed           # active | completed | abandoned (default: completed)
    userIdentifier: "sarah@example.com"
    hoursAgo: 2                 # how far back to timestamp messages (default: 1)
    messages:
      - role: user
        content: "Hi, I want to check on my order ORD-1234."
      - role: assistant
        content: "Let me look that up for you."
    feedback:                   # optional
      verdict: good             # good | bad
      comment: "Very helpful"
      source: customer          # customer | support | system
    links:                      # optional
      - url: "https://store.acme.com/orders/1234"
        title: "Order ORD-1234"
        resourceType: "order"

How to Prepare Files for a New Organization

When the user describes their organization, follow this process:

  1. Gather requirements: What connectors do they need? What agents? What workflows (SOPs)?
  2. Create the directory: anywhere the user wants (e.g., ~/onboarding/my-customer/)
  3. Write files in dependency order: org → users → secrets → connectors → agents → sops → guardrails → evals → sessions
  4. Validate with dry-run: cd modelguide-api && bun run src/cli/mg.ts setup /path/to/dir --dry-run
  5. Run the import: add --skip-secrets for testing, or run without flags for production

For the full schema reference with every field, type, and constraint, read references/schemas.md. For available connector catalog entries and SOP templates, read references/catalog.md. For a complete working example (Acme Corp), read references/examples.md.

Running the CLI

All commands must be run from the modelguide-api/ directory:

bash
cd modelguide-api
bun run src/cli/mg.ts <command> [options]

Running against Railway (from your local machine):

bash
cd modelguide-api
railway run --service api -- sh -c \
  'DATABASE_URL=postgresql://modelguide_app:$APP_DB_PASSWORD@$POSTGRES_TCP_PROXY_DOMAIN:$POSTGRES_TCP_PROXY_PORT/$PGDATABASE \
   bun run src/cli/mg.ts setup /path/to/my-org/ --skip-secrets'

railway run injects all env vars (secrets, encryption keys, etc.). DATABASE_URL is overridden with the public TCP proxy since the private hostname isn't reachable locally. Requires TCP proxy vars from railway/DEPLOY.md step 6.

Common Patterns

Add a single agent to an existing org:

bash
bun run src/cli/mg.ts add-agents --org acme name="New Agent" slug=new-agent modality=voice

Compile only one agent:

bash
bun run src/cli/mg.ts compile-agents --org acme --agent acme-voice-agent

Import SOPs without activating (review first): Set status: draft in sops.yaml, import, review in dashboard, then activate manually.

Re-run after fixing a YAML error: Safe to re-run — duplicates are skipped. Only new entities get created.

© modelguide, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (references) in .claude/skills/mg-cli of modelguide/modelguide.

  • SKILL.md
  • references/catalog.md
  • references/examples.md
  • references/schemas.md

Open the folder on GitHubat commit 554caa0

Compare with similar skills

Mg CLI next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Mg CLI compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Mg CLI this skillmodelguide/modelguide108—~3.5kAutomated safety check: PassMIT
Agentsop Crewaiagentsope/SkillAlchemy457—~4.8kAutomated safety check: PassMIT
Agentsop Dspyagentsope/SkillAlchemy457—~7kAutomated safety check: PassMIT
Agentsop Idempotent Ingestionagentsope/SkillAlchemy457—~6.8kAutomated safety check: PassMIT
Agent Creatoraiskillstore/marketplace4301 repos~4.8kAutomated safety check: PassNone
Agentsop Difyagentsope/SkillAlchemy457—~5.4kAutomated safety check: NotesMIT

Similar skills

  • Agentsop Crewai

    agentsope/SkillAlchemy

    SOP for building multi-agent systems with CrewAI — role-based collaboration, sequential/hierarchical processes, Flows, memory, delegation.

    457 GitHub stars~4.8k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed
  • Agentsop Dspy

    agentsope/SkillAlchemy

    Operating SOP for DSPy (Stanford NLP) — the declarative framework for "programming, not prompting" language models.

    457 GitHub stars~7k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed
  • Agentsop Idempotent Ingestion

    agentsope/SkillAlchemy

    Re-ingest-correctness SOP for production RAG. An agent skill from agentsope/SkillAlchemy.

    457 GitHub stars~6.8k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed
  • Agent Creator

    aiskillstore/marketplace

    Creates specialized AI agents with optimized system prompts using the official 4-phase SOP methodology from Desktop .claude-flow, combined with evidence-based prompting techniques and Claude Agent…

    430 GitHub starsUsed in 1 repo~4.8k tokens
    AI & LLM EngineeringAuto-check passed
  • Agentsop Dify

    agentsope/SkillAlchemy

    SOP for building LLM applications on Dify — visual workflow + chatflow + agent + RAG knowledge base + plugin marketplace + observability, self-hostable.

    457 GitHub stars~5.4k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check: notes
  • Agentsop Framework Selection

    agentsope/SkillAlchemy

    Neutral, framework-agnostic decision tree for project kickoff: "which agent / RAG / LLM framework should I reach for?" Synthesizes the ecosystem sections of 7 landmark-project SOPs (LangGraph…

    457 GitHub stars~5.8k tokensUpdated 1 mo ago
    AI & LLM EngineeringAuto-check passed

More from modelguide/modelguide

  • Mg Connector

    modelguide/modelguide

    A skill your agent uses when the user asks to add a connector, create a connector, implement a connector for a service, add a tool to a connector, update a connector tool, or asks about connector…

    108 GitHub stars~4.5k tokensUpdated 3 mo ago
    Auto-check passed
  • Mg Integrity Check

    modelguide/modelguide

    Use before committing significant code changes to verify project integrity.

    108 GitHub stars~2.4k tokensUpdated 3 mo ago
    Auto-check: notes
  • Mg Local DB Restore

    modelguide/modelguide

    Trigger phrases - "reset local db", "recreate local postgres", "restore dump to local", "reset local database", "load backup locally"

    108 GitHub stars~645 tokensUpdated 3 mo ago
    Auto-check passed
  • Mg Railway DB Backup

    modelguide/modelguide

    Backup a PostgreSQL database from a Railway environment. An agent skill from modelguide/modelguide.

    108 GitHub stars~534 tokensUpdated 3 mo ago
    Auto-check passed
  • Mg Railway DB Restore

    modelguide/modelguide

    Restore a PostgreSQL database backup to a Railway environment.

    108 GitHub stars~562 tokensUpdated 3 mo ago
    Auto-check passed
  • Mg Railway Deploy All

    modelguide/modelguide

    Deploy all ModelGuide services (API, UI, LB) to a Railway environment.

    108 GitHub stars~667 tokensUpdated 3 mo ago
    Auto-check passed

Questions about Mg CLI

What does Mg CLI do?

Generate YAML configuration files and run CLI commands to onboard organizations into ModelGuide. Mg CLI is an agent skill from modelguide/modelguide. Generate YAML configuration files and run CLI commands to onboard organizations into ModelGuide.

When should I use Mg CLI?

Mg CLI fits situations like: the user asks to set up an org; prepare onboarding YAML; provision a customer; anything related to the mg CLI tool.

How do I install Mg CLI in Claude Code?

Run `npx skills add modelguide/modelguide --skill mg-cli -a claude-code`. Or copy the skill folder (.claude/skills/mg-cli in modelguide/modelguide) into .claude/skills/mg-cli in your project. Claude Code loads it when a task matches its description.

How do I install Mg CLI in Codex?

Run `npx skills add modelguide/modelguide --skill mg-cli -a codex`. Or copy the skill folder (.claude/skills/mg-cli in modelguide/modelguide) into .agents/skills/mg-cli in your project. Codex loads it when a task matches its description.

Can I use Mg CLI in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add modelguide/modelguide --skill mg-cli -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/mg-cli, .gemini/skills/mg-cli, .github/skills/mg-cli and .opencode/skills/mg-cli in your project.

What does Mg CLI need to run?

Going by SKILL.md and its folder, Mg CLI needs the command-line tools its instructions call (bun and railway) and credentials named APP_DB_PASSWORD.

Does Mg CLI access the network?

SKILL.md names 1 domain. In commands or code: store.acme.com; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Mg CLI safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Mg CLI use?

Mg CLI is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Mg CLI use?

About 3.5k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 9.1k tokens, read only when the agent opens those files.

What are the alternatives to Mg CLI?

Skills that share tags, products or a category with Mg CLI: Agentsop Crewai (agentsope/SkillAlchemy, 457 stars), Agentsop Dspy (agentsope/SkillAlchemy, 457 stars), Agentsop Idempotent Ingestion (agentsope/SkillAlchemy, 457 stars) and Agent Creator (aiskillstore/marketplace, 430 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Mg CLI?

modelguide (a GitHub organization) maintains it in modelguide/modelguide, which has 108 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on June 20, 2026.

Source: modelguide/modelguide on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.